生理药代动力学(physiologically based pharmacokinetic，PBPK)模型已经被广泛用于预测药物的吸收、分布、代谢和排泄等特性，而基于机器学习(machine learning，ML)和人工智能(artificial intelligence，AI)可以和PBPK模型进行深度融合，从而加快PBPK的预测速度和提高其预测质量，进一步加快药物研发进展。本文介绍了机器学习和人工智能在药代动力学中的应用，对基于机器学习和人工智能的生理药代动力学模型研究进展进行综述，并分析了机器学习和人工智能应用的局限性以及其应用前景和展望。
Research progress of artificial intelligence combined with physiologically based pharmacokinetic models
Physiologically based pharmacokinetic(PBPK)models have been widely used to predict various stages of drug absorption,distribution,metabolism and excretion.Models based on machine learning(ML)and artificial intelligence(AI)can provide better ideas for the construction of PBPK models,which can accelerate the prediction speed and improve the prediction quality of PBPK.ML and AL can complement the advantages of PBPK model to accelerate the progress of drug research and development.This review introduces the application of machine learning and artificial intelligence in pharmacokinetics,summarizes the research progress of physiological pharmacokinetic models based on machine learning and artificial intelligence,and analyzes the limitations of machine learning and artificial intelligence applications and their application prospects and prospects.
physiologically based pharmacokinetic modelartificial intelligencemachine learningpharmacokineticspharmaceutical toxicologydrug-drug interaction
万方数据
维普
